Replace loguru with default logging (#170)

* Replace loguru with default logging

* Remove loguru from tests

Author: justin13601

pyproject.toml

@@ -21,7 +21,6 @@ dependencies = [
     "polars >= 1.0.0, <= 1.17.1",
     "bigtree == 0.18.*",
     "ruamel.yaml == 0.18.*",
-    "loguru == 0.7.*",
     "hydra-core == 1.3.*",
     "pytimeparse == 1.1.*",
     "networkx == 3.3.*",

src/aces/__main__.py

@@ -1,16 +1,17 @@
 """Main script for end-to-end cohort extraction."""
 
+import logging
 import sys
 from importlib.resources import files
 
 import hydra
 import polars as pl
 import pyarrow as pa
 import pyarrow.parquet as pq
-from loguru import logger
 from meds import label_schema, prediction_time_field, subject_id_field
 from omegaconf import DictConfig
 
+logger = logging.getLogger(__name__)
 config_yaml = files("aces").joinpath("configs/_aces.yaml")
 
 if len(sys.argv) == 1:
@@ -136,12 +137,9 @@ def main(cfg: DictConfig) -> None:
     from datetime import datetime
     from pathlib import Path
 
-    from hydra.core.hydra_config import HydraConfig
     from omegaconf import OmegaConf
 
-    from . import config, predicates, query, utils
-
-    utils.hydra_loguru_init(f"{HydraConfig.get().job.name}.log")
+    from . import config, predicates, query
 
     st = datetime.now()
 

src/aces/config.py

@@ -4,6 +4,7 @@
 from __future__ import annotations
 
 import dataclasses
+import logging
 import re
 from collections import OrderedDict
 from dataclasses import field
@@ -15,7 +16,6 @@
 import polars as pl
 import ruamel.yaml
 from bigtree import Node
-from loguru import logger
 
 from .types import (
     ANY_EVENT_COLUMN,
@@ -26,6 +26,8 @@
 )
 from .utils import parse_timedelta
 
+logger = logging.getLogger(__name__)
+
 
 @dataclasses.dataclass
 class PlainPredicateConfig:

src/aces/constraints.py

@@ -1,10 +1,13 @@
 """Contains utilities for validating that windows satisfy a set of constraints."""
 
+import logging
+
 import polars as pl
-from loguru import logger
 
 from .types import ANY_EVENT_COLUMN
 
+logger = logging.getLogger(__name__)
+
 
 def check_constraints(
     window_constraints: dict[str, tuple[int | None, int | None]], summary_df: pl.DataFrame

src/aces/extract_subtree.py

@@ -1,14 +1,16 @@
 """This module contains the functions for extracting constraint hierarchy subtrees."""
 
+import logging
 from datetime import timedelta
 
 import polars as pl
 from bigtree import Node
-from loguru import logger
 
 from .aggregate import aggregate_event_bound_window, aggregate_temporal_window
 from .constraints import check_constraints
 
+logger = logging.getLogger(__name__)
+
 
 def extract_subtree(
     subtree: Node,

src/aces/predicates.py

@@ -1,9 +1,9 @@
 """This module contains functions for generating predicate columns for event sequences."""
 
+import logging
 from pathlib import Path
 
 import polars as pl
-from loguru import logger
 from omegaconf import DictConfig
 from polars.exceptions import ColumnNotFoundError
 
@@ -15,6 +15,8 @@
     START_OF_RECORD_KEY,
 )
 
+logger = logging.getLogger(__name__)
+
 
 def direct_load_plain_predicates(
     data_path: Path, predicates: list[str], ts_format: str | None

src/aces/query.py

@@ -3,16 +3,18 @@
 It accepts the configuration file and predicate columns, builds the tree, and recursively queries the tree.
 """
 
+import logging
 
 import polars as pl
 from bigtree import preorder_iter
-from loguru import logger
 
 from .config import TaskExtractorConfig
 from .constraints import check_constraints, check_static_variables
 from .extract_subtree import extract_subtree
 from .utils import log_tree
 
+logger = logging.getLogger(__name__)
+
 
 def query(cfg: TaskExtractorConfig, predicates_df: pl.DataFrame) -> pl.DataFrame:
     """Query a task using the provided configuration file and predicates dataframe.

src/aces/utils.py

@@ -1,15 +1,15 @@
 import io
-import os
+import logging
 import sys
 from collections.abc import Generator
 from contextlib import contextmanager
 from datetime import timedelta
 
-import hydra
 from bigtree import Node, print_tree
-from loguru import logger
 from pytimeparse import parse
 
+logger = logging.getLogger(__name__)
+
 
 def parse_timedelta(time_str: str = None) -> timedelta:
     """Parse a time string and return a timedelta object.
@@ -66,9 +66,3 @@ def log_tree(node: Node) -> None:
     with capture_output() as captured:
         print_tree(node, style="const_bold")  # This will print to the captured StringIO instead of stdout
     logger.info("\n" + captured.getvalue())  # Log the captured output
-
-
-def hydra_loguru_init(filename: str) -> None:
-    """Must be called from a hydra main!"""
-    hydra_path = hydra.core.hydra_config.HydraConfig.get().runtime.output_dir
-    logger.add(os.path.join(hydra_path, filename))

tests/test_meds.py

@@ -5,13 +5,13 @@
 
 root = rootutils.setup_root(__file__, dotenv=True, pythonpath=True, cwd=True)
 
+import logging
 import tempfile
 from io import StringIO
 from pathlib import Path
 
 import polars as pl
 import pyarrow as pa
-from loguru import logger
 from meds import label_schema, subject_id_field
 from yaml import load as load_yaml
 
@@ -28,6 +28,7 @@
 except ImportError:
     from yaml import Loader
 
+logger = logging.getLogger(__name__)
 pl.enable_string_cache()
 
 TS_FORMAT = "%m/%d/%Y %H:%M"

tests/utils.py

@@ -4,14 +4,16 @@
 
 root = rootutils.setup_root(__file__, dotenv=True, pythonpath=True, cwd=True)
 
+import logging
 import subprocess
 import tempfile
 from pathlib import Path
 
 import polars as pl
-from loguru import logger
 from polars.testing import assert_frame_equal
 
+logger = logging.getLogger(__name__)
+
 
 def run_command(script: str, hydra_kwargs: dict[str, str], test_name: str, expected_returncode: int = 0):
     command_parts = [script] + [f"{k}={v}" for k, v in hydra_kwargs.items()]